update README.md

README.md

@@ -1,10 +1,25 @@
 # sompy
-numpy based SOM (Self Organizing Map) Library
+a numpy based SOM (Self Organizing Map) Library
 
 
 ## Description
 
 ### Demo
+    from sompy import SOM
+    import numpy as np
+    import matplotlib.pyplot as plt
+
+
+    input_data = np.random.rand(10000, 3)
+    output_shape = (40, 40)
+    som = SOM(output_shape, input_data)
+    som.set_parameter(neighbor=0.1, learning_rate=0.2)
+
+    output_map = som.train(10000)
+
+    plt.imshow(output_map,
+               interpolation='none')
+    plt.show()
 ![alt tag](https://github.com/ttlg/sompy/blob/master/examples/sample_color.png)
 
 ## Requirement

examples/sample_color.py

@@ -1,18 +1,15 @@
-import numpy as np
 from sompy import SOM
+import numpy as np
 import matplotlib.pyplot as plt
 
 
-N = 40
-
-input_layer = np.random.rand(10000, 3)
-output_shape = (N, N)
-
-som = SOM(output_shape, input_layer)
+input_data = np.random.rand(10000, 3)
+output_shape = (40, 40)
+som = SOM(output_shape, input_data)
 som.set_parameter(neighbor=0.1, learning_rate=0.2)
 
-output_layer = som.train(10000)
+output_map = som.train(10000)
 
-plt.imshow(output_layer,
+plt.imshow(output_map,
            interpolation='none')
 plt.show()

sompy/sompy.py

@@ -66,8 +66,10 @@ def _neighbourhood(self, t):
         initial = max(self.shape) * self._param_neighbor
         return initial*np.exp(-t/self._life)
 
-    def train(self, n):
-        for i in range(n):
+    def train(self, iterate=None):
+        if not iterate:
+            iterate = self.input_num
+        for i in range(iterate):
             r = rand.randint(0, self.input_num)
             data = self.input_layer[r]
             win_idx = self._get_winner_node(data)